econstor
A Service of
zbw
Make Your Publications Visible.
Leibniz-Informationszentrum
Wirtschaft
Leibniz Information Centre
for Economics
Möllenberg, Antje
Working Paper
Internet auctions in marketing: The consumer
perspective
Arbeitspapier // Technische Universität Braunschweig, Institut für Marketing, No. 03/02
Provided in Cooperation with:
Technische Universität Braunschweig, Institute of Marketing
Suggested Citation: Möllenberg, Antje (2003) : Internet auctions in marketing: The consumer
perspective, Arbeitspapier // Technische Universität Braunschweig, Institut für Marketing, No.
03/02, ISBN 3933628482
This Version is available at:
http://hdl.handle.net/10419/54775
Standard-Nutzungsbedingungen:
Terms of use:
Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen
Zwecken und zum Privatgebrauch gespeichert und kopiert werden.
Documents in EconStor may be saved and copied for your
personal and scholarly purposes.
Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle
Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich
machen, vertreiben oder anderweitig nutzen.
You are not to copy documents for public or commercial
purposes, to exhibit the documents publicly, to make them
publicly available on the internet, or to distribute or otherwise
use the documents in public.
Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen
(insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten,
gelten abweichend von diesen Nutzungsbedingungen die in der dort
genannten Lizenz gewährten Nutzungsrechte.
www.econstor.eu
If the documents have been made available under an Open
Content Licence (especially Creative Commons Licences), you
may exercise further usage rights as specified in the indicated
licence.
Antje Möllenberg
Internet auctions in marketing:
the consumer perspective
AP-Nr. 03/02
Technical University ofBraunschweig
College of Business Administration
Department of Marketing
ISBN 3-933628-48-2
Contents:
Page:
Abstract ............................................................................................. 1
1
Introduction ................................................................................. 2
2
A consumer-oriented framework of internet auctions ................... 2
2.1
Defininition of internet auctions .......................................... 2
2.2 Business models .................................................................. 4
2.3
3
Shopping at an auction site .................................................. 6
Previous research on auction customers ..................................... 10
3.1
Auction theory .................................................................. 10
3.2 Consumer research............................................................ 11
3.3
Internet user research ........................................................ 13
3.4 Conclusion from theories .................................................. 15
4
Empirical study of auction customers......................................... 15
4.1
Method .............................................................................. 15
4.2 Results .............................................................................. 16
4.2.1 User types ............................................................... 16
4.2.2 Auction usage ......................................................... 17
4.2.3 Bidding behavior .................................................... 20
4.2.4 Future marketing potential ...................................... 22
5
Discussion ................................................................................. 26
References ....................................................................................... 28
Author .............................................................................................. 30
1
Abstract
Internet auctions for consumers are among the lnost popular and lnost
successful business models in electronic COlnmerce. Research so far,
however, has focused on prerequisites and consequences of auctions
as a marketing instrument of suppliers. Even though it is a key success
factor from a lnarketing perspective, the delnand side has not inspired
sünilar attention.
This paper focuses on the attitudes, lnotives, and behavior of auction
custolners. It shows why current beliefs about bidder characteristics
are lnyths. Taking these lnisconceptions as a starting point, the existence of an experiential and a praglnatic type of auction customer is
proposed. An explorative empirical study looking for the characteristics of both types of auction customers is described.
Results indicate that less than half of auction shoppers in the study are
experiential oriented. Except substantial additional demand concerning technological and emotional qualities of auctions these shoppers
do not differ dramatically from pragmatic oriented shoppers. Both
types are open-lninded towards further development of consumer
auctions to commercial marketplaces. Business models of auctioneers
and suppliers should concentrate on the basic utility of the auction
algorithlTI by facilitating individual matchmaking instead of pursuing
costly additional utility by promoting the entertainment value of
auctions.
2
1
Introduction
Internet auctions for consumers are one of the web' s biggest success
stories. Even after the initial euphoria about the "new economy" has
been replaced by more realistic appraisals of e-commerce, the auction
industry is regarded as a paragon of the net-based economy (Porter
2001, p. 67). Eliciting consumers' enthusiasm from the very beginning, a variety of most singular goods have been auctioned on the
web, such as a vacation at Bahnoral Castle, pop diva Madonna's bra,
or debris of the Berlin Wall. The fact that auctions are increasingly
used as a platform for everyday goods such as CDs, cOlnputers, books,
and clothes is somewhat overshadowed by such unique, if spectacular
auctions.
Consulner encounter internet auctions either as auctions from consumer to consumer (C2C) or as auctions from business (auctioneers or
suppliers) to consumers (B2C). The "2C" in both suggests the term
"consumer auction". Early C2C auctions were often dismissed by
economists as "flea markets" (Schrage 2000, p. 92). B2C auctions
have sparked considerably more interest in researchers and marketers.
Still, this attention has focused on the supply side and neglected the
customer. From a marketing perspective this is a serious omission, as
consumer orientation or indeed any business Inodel must be based on
valid information about the customer. The intense and widespread involvement of consumers in internet auctions justifies a closer look at
their motivation, attitudes, and behavior. From this analysis insights
about the preconditions and consequences of using auctions in marketing may be gained.
2
A consumer . . oriented framework of internet auct ..
ions
2.1 Defininition of internet auctions
In business administration, auctions are defined as market institutions
that take place as sporadic, real-life events, requiring the physical
3
presence ofparticipants and goods. These characteristics obviously do
not apply to auctions in the virtual market space. Interestingly, in Germany legal definitions of auctions do not exist. Moreover, legislation
such as it is, is applicable to traditional auctions only. As they do not
fit into the legal scheme, internet auctions have caused severe problems, many but not all of which have been amended by recent EU
guidelines on e-commerce. A persistent, big obstacle as far as marketing is concerned is the lack of confidence in legal aspects of auctions
in the minds ofboth custoluers and suppliers.
The definition of auction theory is more appropriate. Auctions are
viewed as institutionalized methods for price formation, in which the
allocation of resources is determined by an explicit algorithm based on
the bids of market participants (McAfee/McMiIlan 1987, p. 701). The
active, dominant role of the demand side is crucial to the auction process (Cassady 1967, p. 8). Internet auctions are best described as a virtual market institution relying on internet services (especially WWW
and e-mail) to implement central (dynamic, bidder-driven price formation and allocation) as weIl as peripheral (catalog or bidder's
register) auction features.
Internet auctions are modeled closely after traditional auctions. The
internet helps to overcome temporal and spacial restrictions of traditional auctions and brings about a grave reduction in trans action costs.
Generally, auctions are indicated in situations of market failure, e.g.
when there is incomplete competition or uncertainty about prices. This
may be the case when unique goods are offered, when the variability
of prices is high, or when large amounts of goods have to be distributed very quickly. However, while classic auctions are restricted either
to expensive, luxurious goods such as antiques or art, or to large
amounts of identical merchandise such as produce, on the "virtual auction floor" the auction mechanism can be applied to every kind of
product or user. Thus, with auctions the internet has enabled C2C
electronic commerce on a wide scale (Lührig/Dholakia 2002, p. 120).
One consequence of the heterogenity of auction definitions is the
ambiguity of the term "auction" on different semantic levels. On the
macroscopic level, an auction is an organized market event that con-
4
sists of multiple single auction sales. In this sense that recalls the definition of business administration, auctions are comparable to a shop
offering a variety of goods. On the microscopic level, an auction is a
specific instantiation of the auction algorithm and its parameters,
leading to a single auction sale. This interpretation corresponds to the
view of game theory on actions. And finally, internet auctioneers
sometimes also are referred to as "auctions". Thus, "internet auction"
may refer to a certain auction sale (e.g. a single eBay auction for a
vintage teddy bear), an organized auction event, where several items
are being auctioned (e.g. all offers on the eBay platform at a given
point oftime), or an internet auctioneer (e.g. the eBay company). This
plethora of meanings may cause serious misunderstandings.
2.2 Business models
Internet auctioneers adopt one of two different business models, which
entail opposing marketing concepts (table 1).
Store auctions (Elliott 2000, p. 2f.) realize the B2C consumer auction
model by offering new merchandise and branded goods. To this end
they cooperate with brand manufacturers and logistic services. Often
they act as retailers by purchasing, stocking, and delivering goods.
Short-termed auctions of just a few minutes duration, often conducted
in real time, and presented by a human moderator mimic the glamorous atmosphere of real auctions. This so-called auctainment is meant
to attract entertainment-seeking consumers. A parallel is the brand
strategy: in saturated markets, consumer products and services are frequently augmented by additional emotional utility in order to distinguish a brand from its competitors. Two rationales underlie this approach (1) attracting customers by auctioning well-known brands, and
(2) attempting to stand out in the saturated e-commerce market with
its low entry barriers for competitors and many similar shopping and
auction sites just "one click away" .
The earliest example of the auctainment approach was US auction
pioneer Onsale, founded in May 1995 (Lucking-Reiley 2000, p.
228f.). The model was adopted by many German pioneers. Examples
are Ricardo (founded late 1998), 12snap (a pioneer of mobile auc-
5
tions) and Primus auctions (an e-commerce branch of Metro Holding,
the largest trading company in Germany). After an initial period of
growth and enthusiasm however, nearly every B2C auction went out
ofbusiness or re-Iaunched their sites as C2C auctions.
Open auctions implement a C2C consumer auction Inodel, although
many of their representatives such as eBay currently are incorporating
B2C activities as weIl. Open auctions list used goods, employ long
term auctions between 3 and 14 days, and dispense with human auctioneers. Their users are assumed to be motivated by "bargains" rather
than by experiential aspects of auctions. The most popular example is
eBay (founded September 1995). The company entered the German
market as a follower by taking over its German imitator Alando. de in
June 1999, when Alando had been on the market for just three months.
Although competition is strong, C2C auctions thrive and aim for the
B2C market.
Models differ greatly in the way they target auction custolners. The
auctainment model as represented by early Ricardo attempts to generate exciting and entertaining auction events, while the matchmaking
model as represented by early eBay facilitates finding and closing the
best possible auction deal. In doing so, the auctainment model emphasizes the auction process, making the consumers' involvement in
the bidding an important success factor. The matchmaking model on
the other hand focuses on the outcome of the auction.
Obviously the auctainment model primarily targets a specific market
segment, the thrill-seeking auctainers. The appeal of novelty and variety as an end in itself as well as enjoyment ofthe bidding competition
are characteristics of this type of auction participant. The type of user
that is addressed by the matchmaking model is much less obvious.
negotiating a deal at a personal price. Thus, it may safely be assumed
that the prevailing user type here is price, or rather value oriented.
Although the assumptions underlying both auction models offer some
face validity, neither has been studied empirically. Very few studies of
auction users actually exist. Most of what is assumed to be true about
6
auction users results from the portrayal of auctions and their users in
popular media and the professional field and is no more than a myth
(Herschlag/Zwick 2000). From SOlne characteristics of auctions such
as the risk they entail, the enticing price mechanism, the glamorous
atmosphere of real auctions, and their unsuitability for everyday
shopping, it is concluded that auction users must be risk-seeking,
thrifty, and easy to snare by unique and even overpriced offers, if only
they are presented in a stimulating atmosphere. This implicit personality theory (Cronbach 1955; Pervin 1978) yields a good starting point
for empirical research on auction users but should not be the foundation on which to build a whole industry. Customizing auction marketing for auctainers or for bargainers may both fail: Auctainers are hard
to please, because bew attractions must be offered continuously or the
auction experience will cease to be entertaining; bargainers constitute
only about 8 % of active internet users (McKinsey 2000), which may
not suffice to sustain a business model in the long run.
Transaction area
C2C
B2C
Business model
Open Auction
Store Auction
Example
eBay,
Yahoo, hood
Onsale, Ricardo,
Primus, Atrada, 12snap
Differentiation by
product range
Used goods
New consumer goods,
brand products
Long term auctions
Short term, real time,
and event auctions
Best bargain
Auction experience
Matchmaking
Auctainment
Bargainers
Auctainers
Differentiation by
auction form
Differentiation by
customer utilitv
Market
positioning
Target market
Table 1: Business models of internet auctioneers
2.3 Shopping at an auction site
Auctions involve three players with distinctive roles, motives, benefits, and risks (figure 1). The auctioneer evokes bids from the auction
floor, repeats them aloud to bidders and knocks the item down to the
7
highest bidder. In internet auctions, these role is usuaHy taken by the
auction software; only the auctainment model featured human auctioneers. Because of the seller's passivity in traditional auctions, the roles
of seHer and auctioneer often are not delnarcated (see 3.1). In internet
auctions the seHer is more active. The bidder plays the most prominent role in auctions, as auctions are determined by bidder activity
(Cassady 1967, p. 8; Smith 1989, p. 174f.).
Internet auctioneer
(market operator)
Bidder
(demand side)
Immediate market
transactions
Seiler
(supply side)
Figure 1: The tripolar structure of internet auctions
The auction process consists of three successive phases: before, during, and after the sale (figure 2). Most discussions of internet auctions
focus on the price determination process. However, each phase requires active participation and implies its own marketing chaHenges.
Exaggerated emphasis on the bidding process oversimplifies auctions
and fosters misconceptions about auction users by concentrating on
the most variable and exciting aspects of internet auctions.
e
In the pre-sales phase, bidders locate and evaluate attractive offers.
SeHers design and promote sales offers by selecting adequate
values for the auction parameters (starting price and duration), and
by providing the buyer with information (pieture and verbal description of the item). The pre-auction phase ends with the bidder's
decision to bid. Auctioneers support this stage by providing auction
tools and counseling bidder and seHer.
8
" The safes phase starts with the auction phase. Bidders take a very
active part, continuously watching the auction and reappraising
their valuation as others place bids, and alternative offers turn up.
SeIlers must content themselves with a passive role as they cannot
influence the auction once the item goes online in the auction system. I Bidding is terminated according to an acceptance rule. Long
term auctions usually end after a pre-defined length of time; short
term auctions end when no bidding activity has been registered for
some time. Auctioneers support this phase by intelligent software
tools called bidding assistants. The second part of the sales phase
consists of payment and delivery. Auctions can get extremely tedious at this stage, especially if many auctions must be managed at
the same time. Many tools provided by auctioneers facilitate trading in this transaction phase, e.g. escrow or payment services.
" After the auction the activity shifts to seIler and auctioneer. The
seIler prepares future sales. He is also answerable to guarantee
claims. Auctioneers support this phase by custoluer-relationship
and cross-selling tools. Auctions are naturally biased towards single
transactions, so it is important though difficult to establish longterm relations between customer and seller as well as customer and
auctioneer.
Substitutes for auctions from the shoppers ' view may be any shopping
form that offers comparable advantages to customers. Although no
alternative has all distinctive characteristics of internet auctions, many
substitutes can be identified both online and offline. In the C2C
domain there are classified ads and flea markets (Bearchell 1999;
Porter 2001, p. 67) as weIl as P2P online exchanges for deals with
other consumers. In the B2C domain discounters and factory outlets
offer attractive deals, mail order businesses and online shops conduct
business from a virtual distance, and unique items at a bargain price
can be fbund at real auctions or specialized retailers. But only internet
auctions combine all of these qualities into one on a regular basis.
1 German internet auctions do not even offer areserve price option.
9
Pre-sales phase
Sales phase
r _.... _.. _- ---- .................... _.. _...... ,.- ---- .......... --- -_ .............. _..
Pre-auction phase
• Search & select
• Evaluation
• Decision to
participate
:;~os~ ~e~ ~t
:
• Advertising and
promotion
:u
Q)
c:
o
U
:::l
«
• Presentation
• Bundeling
• Counseling
••...•...
Auction phase
• Bidding
• Watching
• Reappraisal
-~~
~: ~:
After-sales phase
.... ........... --- -_ .. -- ...... _.. -- -"-"r" .................................................... ..
After-auction phase I
• Contact seiler
'. • Payment
: : . Contral delivery
E:JwatChin
ii:
g
After-auction phase 11
• Use
• Re-seil
;~~~~\~~~ing : :g~~s selling
p a y m e n t s . Guarantee
• Shipping
• Auction algorithm
• Support bidding
(e.g. software
agents)
• many services
facilitating auction
management (e.g.
escrow service)
• Tutorials
• Cross selling
• CRM
~~~:i~.t~~~......... ..~!~~i~~.~.~:~~~!~~.~:.•... ~.~~~~~~.~.~.~I~!~r:::~.t.E ...~~~?~~:.:~~~~:.~.~I::~._
Figure 2: Transaction phases in internet auctions
The price mechanism also serves as a criterion for finding alternatives.
In the offline world consumers normaHy encounter fixed-price B2C
commerce. Counter examples are rare; only the oriental bazaar often
serves as a metaphor for internet auctions (e.g. Lührig/Dholakia 2002,
p. 113). For SOlne expensive consumer goods (e.g. cars) , individual
negotiation of contract details also is common. Recently, German law
has permitted individual bargaining in any consumer directed sale,
having regulated price negotiations very tightly before; however,
haggling at the baker's or in the shopping center is inconvenient and
impracticable for both market sides and because of such trans action
costs does not happen in reality. C2C shopping such as classified ads
or flea markets do not impose such restrictions and thus offer adjusted
prices as weH, though at much higher trans action costs than internet
auctions. In the online world, computer-implemented pricing algorithms enable other forms of consumer oriented price determination,
e.g. co-shopping, name-your-price price-seeking, bilateral negotiation,
or exchanges. Again, internet auctions are unique from the consumers'
viewpoint, as they are the only price building mechanism to incorporate competitive pricing on the demand side and leave the supplier with
a take-it-or-Ieave-it option.
10
3
Previous research on auction customers
Three fields of research apply to the study of consumer behavior at
internet auctions, classic auction theory, consumer research, and internet user studies. These theories are now examined for evidence supporting the existence of two separate types of auction shoppers.
3.1 Auction theory
In his seminal paper, Vickrey (1961) laid down the cornerstones of
auction theory, which were to dominate the field in the following
years. He outlined two basic auction models and compared them in
terms of effectiveness and outcome for the seHer. Since that time,
auction theory has had a normative, supply-side emphasis (without
distinguishing between the seHer and the auctioneer), giving implications as to how auctions should be designed in order to maxünize the
seHer's profit. The heyday of auction theory was in the 1980s, centering around the major contributions by Milgrom/Weber (1982) and
McAfee/McMillan (1987), both with the traditional seHer focus. During forty years of auction theorj, the buyer's point of view has been
addressed in a single paper (Matthews 1987), and even that at closer
scrutiny turns out to be written from the seHer' s perspective.
In auction theory bidders are conceptualized as highly rational individuals. They are assumed to reselnble each other closely regarding
valuations, resources, and behavior. Bidder homogeneity is modeled
by drawing the valuations on which bids are based from a common
distribution without revealing them to the other bidders .
., In the simplest auction model valuations are independent of others'
estimates and are determined by individual preferences. Situations
in which this independent private values model applies are auctions
für coHectibles, for rare or unique items or any good that is intended for personal use only, so that the price can only depend on subjective utility.
• If on the other hand estimates are correlated because the bidders
assume a common value to the good being auctioned, the common
11
values model applies. This is the case for resaleable goods, or any
consumer good with a listed price or store price that is COlnmon
knowledge. Bids may still differ because this COlnmon value often
is unknown and has to be estimated. Table 2 provides an overview
of both basic models.
Model
Independent private values
Common values
Valuation basis
individual. subjective utility.
ex ante known
collective. objective value.
ex ante unknown
Interdependence of
valuations
independent preferences
estimates of objective value
inte rcorre lated
Source of risk
preference uncertainty
quality uncertainty
Bids signalling
other bidders' preferences
unknown true value
Applicability
rare or unique items
personal use
consumer good with listed price
items for resale
Example
e.g. collectibles
e.g. licences. consumer goods
Table 2: Basic models of auction theory
Preferences are exogenous to the model, as are individual differences
in behavior, resources or attitudes. Any diversion in these factors before, during, or after the auction cannot be captured by auction theory.
Also, while classical auctions were being held for homogenous groups
ofbuyers (e.g. flower sellers, antiques or stamp collectors) who often
were chosen and invitated by the auctioneer, the users of internet auctions are extremely heterogenous due to virtually unrestricted access.
So, looking at auction theory does not really help. Still, auction theory
suggests two basic types of auctions (although this distinction is based
on the type of goods) and allows for the inference of two types of
users.
3.2 Consumer research
Two areas of consumer behavior research seem related to auction
buyer behavior: purehase decisions (Kotler/Armstrong 1994, p.
162ff.) and purehase motives driving a consumer' s choice.
12
There are four kinds of purchase decisions which usually go with a
special type of good (Kotler/Annstrong 1994, p. 162ff., p. 278):
GI
GI
GI
GI
Extensive purehase decisions occur when the good to be purchased
is expensive, and the decision is unique. As all alternatives are new
and must be considered, deciding takes a lot of cognitive effort.
Every shopper new to internet auctions or looking for luxurious or
very rare offers (specialty goods) is forced to extensive decisions.
Limited purehase decisions can fall back on a reduced set of evoked
alternatives because of previous shopping experience or a preselecti on of brands. The effort of deciding also is smaller. The type of
good associated with this type of decision making is called shopping goods. As will be shown below, most purchase decisions in internet auctions can be expected to be of this type.
Impulsive purehase decisions are triggered by a stimulus in the
shopping situation, leading to a spontaneous, uncontrollable urge to
buy. The goods most appropriate to these decisions are called
impulse goods. Some internet auctions try to induce this kind of
decision making by live auction events and human moderation
(auctainment model; see 2.2).
Habitual purehase decisions require very little effort. They result
from repeated occurences of any other type of purchase decision
which cause complex cognitive processes to be compiled into
simpler behavioral routines that are executed automatically on
presence of the stimulus. Goods which widely correspond to habitual behavior are called convenience goods.
Auctions provide buyers with several ways to look for offers, fostering at least two ways of decision making: Browsing the auction site
promotes impulsive bidding, whereas search machines constitute a
more rational approach to finding an interesting auction. However,
because each and every auction anew has to be evaluated in terms of
supplier, quality, price level and so on, the auction purchase decision
must always remain a complex decision. This undermines both impulsive behavior as well as the habitualization process. Internet consumer auctions could therefore never really be convenient. This is
especially true if the buyer participates in many auctions simultaneously. Increasing auction experience, acquaintance with the auction
13
platform and previous shopping experience will render limited purchase decisions most probable.
German research on purehase motives has suggested five basic motives: price orientation, experiential orientiation, convenience orientation, brand orientation, and service orientation.
Online shoppers are assulned to be motivated mainly by price, convenience, and brands. Although the original positioning of suppliers as
bargain dealers has probably played an important part in generating
the price pressure predominant in e-commerce today, innovative pricing mechanisms such as auctions do cover an existing need of online
customers. Brand orientation may serve as a means to further differentiate between a me re bargainer who is interested in the cheapest buy,
and the so-called smart shopper, who actively seeks out brands at the
lowest price possible (Esser 1999).
Experiential orientation refers to shopping for motives other than obtaining the product, e.g. diversion from daily routine, self satisfaction,
physical activity, communication, peer group attraction, status, or
pleasure of bargaining (Li/Kuo/Russell 1999). The term recreational
orientation (Tauber 1972) captures such shopping motives more
closely. It has often been argued that the internet is unable to fulfill
experiential demands, concluding that online shoppers should score
lower on items measuring recreational or experiential orientation (e.g.
Loevenich 2002). Empirical evidence on this topic is not straightforward. In an early US study, no differences between frequent onlineshoppers and non-online shoppers conceming recreational orientation
were observed (Li/Kuo/Russell 1999). A more recent German study
found significantly higher recreational as well as price orientation in
online shoppers as compared to non-shoppers (Loevenich 2002). Internet auctions as a comparably emotional type of onIine shopping
may at least partially explain this result.
3.3 Internet user research
Specific studies ab out auction customers are rare. Elliott (2000, p. 39)
differs two types of auction customers: collectors and bargainers.
14
Without further explanation, German researchers have added the
"serious shopper" to these types (Weinhardt/Schmidt 2001), thus
implying that the first two types are not to be taken seriously.
From panel data the German market research company GjK has derived six types of "e-commercers" (i.e. people doing online shopping)
(Spohrer/Bronold 2000). Most prominent were the so-called professionals and practitioners who react strongly to new offers and are interested in attractive and easy-to-use commercial sites. Because of the
potential entertainment value of internet auctions, gameboys/cybergirls as young and brand oriented e-commercers also belong to the
target market.
From MediaMetrix' s US panel, McKinsey has also arrived at six
shopper types. Most important e-commercers are utility-oriented
simplijiers and price oriented bargainers. More than 50 % of visits at
eBay.com are attributed to bargainers looking for good value and
special offers (McKinsey 2000).
Both typologies identify one type that is curious and open-minded but
has been busy more with browsing than with shopping and thus is no
e-commercer (Clicker/GfK..; Surfer/McKinsey).
There are some similarities between these typologies: Each identifies
some types that primarily seek emotional experiences, and atllers that
are rational and goal oriented. This resembles the distinction made by
HoffmanINovak (1996) who suggested two behavioral categories of
internet usage:
• Goal oriented user behavior is characterized by extrinsic motivation, highly structured selective search, and high involvement.
Users are result oriented, i.e. they want to find a solution to a problem or get done with a task. In auctions such behavior occurs when
seeking offers by specified criteria and on definite demand.
• Experiential user behavior is characterized by intrinsic motivation,
unstructured, associative and hedonistic behavior. Users are process
oriented, i.e. carry out the behavior for its own sake. In auctions
this kind of behavior is found as browsing the site without acute
15
demand or as spontanous bidding near the end of an auction. Auctainment is trying to induce and exploit process orientation.
3.4 Conclusion from theories
On examination of implicit user theories, economic theories of auctions and behavioral theories two types of auction shoppers repeatedly
eluerge. This paper therefore proposes the existence of two shopper
types. Based on the corresponding marketing concept the experiential
oriented custoluer is called auctainer. Analogous to the smart shopper
the result oriented customer will be called smart bidder. Each type is
attributed certain preferences, characteristics, and behaviors (table 3).
Attribute
Motive
Information seeking
Intention
Purchase decision
Risk attitude
Shopping
experience
User type
Purpose
Auction model for
brand purchasinQ
Bidding behavior
Marketing concept
Auctainer
experiential/recreational
orientation
browsing; emotional stimuli
process
impulse purchase
risk seeking
Smart Bidder
price orientation,
praamatic orientation
seeking; cognitive stimuli
result
limited purchase
risk averse
unexperienced, new customer
experienced, repeated buyer
experiential/recreational type
personal use, collecting
independent private values
(emotional differentiation)
aggressive bidding
auctainment
bargainer, practitioner
personal use, reselling
common values
(price orientation)
bargain bidding
matchmaking
Table 3: Proposed types of auction customers
4
Empirical study of auction customers
The following data is taken from a doctoral thesis on internet auction
shoppers (Möllenberg 2003).
4.1 Method
The study used an explorative survey design. Shopper types were constructed post-hoc. The differences concerning auction usage, bidding
behavior, and future marketing potential of auctions were then examined using univariate and multivariate statistical procedures.
16
A combination of e-mail survey and web survey seemed appropriate
as internet auctions require their participants to rely heavily on virtual
communication methods anyway. The study was conducted in the
months of May and November as during these periods there are no
major holidays and users are most likely to participate in auctions.
In November 2000, 2,382 auctions at eBay.de were randoluly accessed. Auctions were selected by calling an eBay interface routine returning the list of 100 auctions about to end next. These were then
called one by one (having ended in the meantime), and saved into a
database. Auctions that were illegal, not consumer-oriented, or not
lueant for the Genuan market were excluded from further analysis.
In May 2001, every seIler and highest bidder from these auctions was
contacted by e-mail and asked to participate in a web survey. All in
all, 2,791 addresses were used (1,522 sellers, 1,228 buyers plus 41
more addresses from a practice sampie). After invalid addresses and
refusals to participate were excluded, 2,602 addresses (93.2 %) remained in the sampie. 436 completed web questionnaires (16.8 %)
were returned.
4.2 Results
4.2.1
User types
Subjects classified themselves into four basic user types based on
verbal descriptions. They indicated both the best and the second best
fitting description. Only about one quarter of participants viewed
theluselves as the experiential type. Half said they were the pragmatic
type, and one fifth considered themselves price oriented (figure 3).
17
First choice (n = 422)
i1% ..:::·::A.~:~·~::::::
~~~~~~~~~~~~~~~~~~~------~~~~~~~~~
Second choice (n = 422) ~~~22.8%:~~~
~~~~------~~~~~~~~~
"''''''''''
Experiential type (n = 113) 8.8%
~~~~~~~~~~~~~~~~
Practical type (n = 199)
=31)
Bargainer (n =79)
Surfer (n
~~~~~~~~~~~~~~~~
~~~~~~------~~~~~~~
~~~~--------~------~~~~
0.0%
~ Experiential
D
Practical
50.0%
8
Surfer
100.0%
0
Bargainer
Figure 3: Self categorization into types
Surfers were excluded from further analysis as they were very few and
by definition did not shop at auctions or on the web anyway. The remaining subjects were regrouped by the following rule: All subjects
featuring the experiential type as either first or second choice were regrouped as auctainers. All other subj ects featuring any combination of
price or pragmatic orientation were regrouped as smart bidders. This
process resulted in two comparably sized subsampies with 186 (auctainers) and 205 (smart bidders) subjects.
4.2.2
Auction usage
Smart bidders have been using the internet longer than auctainers
(3.99 vs. 3.63 years; T = -1.616, P = .11). There is no difference in
currentfrequency of internet access, but in access place: auctainers
much more frequently use their ho me access whereas smart bidders
access the net equally or more often from the work place (X2 = 13.91,
df = 2, P < .01). The attitude towards risk was operationalized as a
fixed-price traditional retail buy (certainty equivalent) as opposed to a
risky auction alternative. No significant difference of means was
found: Both types tend to be slightly risk seeking. On closer inspection of the distribution smart bidders turn out to contain more especially risk averse buyers.
18
The analysis of auction experience indicates large interindividual differences within groups rendering means differences mainly insignificant. On the surface, smart bidders bid on (160 vs. 123; T = -0.616) as
weIl as won (33.2 vs. 32.6; T = -0.093) more auctions than auctainers
in the three months preceding the study. They also spent more money
per auction (69.40 DEM vs. 57.00 DEM; T = -1.136). As regards the
type of goods bought, auctainers buy more unique goods and entertainment articles, while smart bidders buy consumer goods. Both customer groups cover a large proportion of their delnand (between 40
and 75 % depending on product category) via auction. FinaIly, the
purpose ofpurehase indicates that auctainers tend to give away their
auction purchases either by resale or as a gift, while Slnart bidders
keep and use them (X2 = 5.58, df= 2, P < .05).
Users were also asked to rate some reasons for and against taking part
in internet auctions. Hardly any differences were found. In the exploratory factor analysis of the reasons for auction participation three
factors were extracted: attractive features of auctions, unattractive
features of other shopping forms, and excfusiveness of auction safes
offers. Auctainers agreed significantly more to the item "bidding is
fun" (figure 4).
19
disagree
agree
completely
completely
no time to look
somewhere else
don't enjoy looking
somewhere else
rare offers
cheap offers
(Ubargains U)
good quality offers
large assortment
bidding/winning
are fun *
,"
IIIIII~'-_',
fair prices
",
- O·
.
• '0
variable prices
Auctainer (n = 186)
-II1II-
Smart Bidder (n
=205)
* p:s; 0,05
Figure 4: Reasons for using auctions
The exploratory factor analysis of reasons against auction participation yielded three factors capturing phase specific problems of auction
transactions. Both types of buyers view the pre-sales phase as most
critical, e.g. quality uncertainty or seHer reliability. Problems during
the sales phase itself are rated second, while the after-sales phase is
seen as fhe smallest cause of difficulties (figure 5).
20
disagree
agree
completely
completely
uncertain about quality
J
reliability of seiler
CII
lose track
of payments
lose track
of deliveries
lose track
of mails
delivery is a problem
unclearness of
assortment
takes too long
uncertain about
competitors' behavior
fraud because of seiler
bidding ("shilling")
uncertainty about
price development
- '0·
Auctainer (n = 186)
-11-
Smart Bidder (n = 205)
Figure 5: Reasons against using auctions
4.2.3
Bidding behavior
Construction of customer types was guided by the idea of separating
experiential oriented customers from all other types. To demonstrate
the success of this attempt bidding behavior is analyzed. Many differences between auctainers and smart bidders can be seen (figure 6). Ex-
21
p1oratory factor analysis returns a five dimensional structure of bidding behavior. The differences can be traced back to two factors: Auctainers score higher on the factor experiential strategy (0.176 vs.
-0.116; T = 2.922, P < .01); smart bidders scorehigher onpragmatic
strategy (-0.129 vs. 0.143; T = -2.749, p < .01). This result may be
taken as va1idity proof of group construction. The other three strategies were named valuation by common value, valuation by independent
value, and an especially conservative cautiousness strategy.
disagree
agree
completely
completely
bid more careful the less
certain objective value
bid more than intended
if involvement high **
bidding is fun **
bid by objective value
(e.g. store price)
result more important
than bidding process (*)
try to find out
other bidders' valuation
don't like to be outbid (*)
am best informed about
value
each bidder has
individual valuation
want to win
at all costs *
sometimes think
I paid too much
bid by personal value
won't let myself
be influenced by others
all value object like I do
• o· Auctainer (n = 186)
-l1li-
(*) p < 0,10 * P < 0,05
Figure 6: Bidding behavior
Smart Bidder (n = 205)
** P ~ 0,01
22
4.2.4
Future marketing potential
The acceptance for new merchandise and for commercial sellers are
important indicators for buyers' affinity to cOlnmercial consumer
auctions (B2C). For both indicators means in the indifference range
were measured; new merchandise was rated slightly lnore favorable
(3.26 on ascale from 1 = rejection to 5 = preference), cOlnmercial
seHers were rated slightly less favorable (2.72). A look at the frequency distribution shows that more smart bidders prefer used merchandise and private seHers, although differences are not statisticaHy
significant.
When asked what features would increase auction adoption, notable
differences between user groups were found (figure 7). Auctainers
showed a lnarked preference for live auctions, fixed-price formats,
and power shopping. In an exploratory factor analysis this bundle of
variables was identified as a separate dimension (innovative sales formats) with significantly higher factor scores of auctainers as compared
to smart bidders (0.202 vs. -0.147; T = 3.390, P < .001).
To auctainers offline information and access are more important than
to smart bidders. In combination with the items "mobile information"
and "mobile access" these items also load on a distinct factor (alternative access). Auctainer score much higher on this factor than Slnart
bidders (0.040 vs. -0.164; T = 2.030, P < .05). Other dimensions of
adoption factors were identified as fulfilment support, additional information, and standard interface features.
23
not important
very
at all
important
Watching function
Feedback function
Chat, community
Live auctions ***
II1II
Fixed price formats
Power shopping **
Additional product
information
Price comparisons
to other dealers
more convenient payments
(e.g. Paypal)
Auction tracking
and handling
Guarantees, insurance
services
improved delivery or
delivery service
II1II)
Mobile alarms
Mobile bidding
Offline catalogue *
Offline access
(e.g. telephone, fax) **
- 0"
Auctainer (n = 186)
* p :::; 0,05
-l1li-
** P :::; 0,01
Smart Bidder (n = 205)
*** P :::; 0,001
Figure 7: Auction adoption factors
To look for perceived alternatives to auctions, subjects were put in a
fictitious forced choice situation (where to shop if there no longer
were such a thing as internet auctions). Auctainers then te nd to prefer
any fonn of e-commerce more than smart bidders (figure 8). Power
shopping is the worst alternative to both groups. The best alternative
for the auctainers is online shopping, for smart bidders brick-and-
24
Inortar retailers. Traditionallnail order rates medium high with both
groups.
This item operationalizes only prefered substitution of auctions by
other shopping forms. The reverse interpretation, substitution of the
alternative shopping fonn by auctions and thus the threat to traditional
distribution channels is not a valid inference. However, the ratings
observed here may serve as a heuristic or a plausible approximation.
no alternative
most approproate
at all
alternative
l1li 0
Flea market
t. /
Classified ads
-
...
o l1li
Newspaper ads
Department store!
specialty retailers
"
\
..
1/l1li
-
IIII~
/.;:
Mailorder
-
IIII~
Power shopping (*)
-
~.·'o
Online shopping
j,/
other Internet
market places
• 0"
/
l1li
l1li0
Auctainer (n = 186)
-l1li- Smart Bidder (n = 205)
(*) p < 0,10
Figure 8: Alternatives to auctions
Today SOlle goods are not typically sold via auctions. In the analysis
of purchasing readiness towards these goods it turned out that both
user types can best imagine the purchase of shopping goods (figure 9).
The least potential is attributed to fresh groceries that indeed seem
rather unsuitable for auction. An exploratory factor analysis additionally returned the dimensions standardizable goods and goods for
everyday use. On this final factor auctainers score significantly higher,
25
i.e. they would tend more to buy everyday consumer goods in auctions. The item services that loaded high both on the factors shopping
goods as well as on goods for everyday use luay be interpreted that
personal services would presumably be bought mainly by auctainers.
disagree
agree
completely
completely
Fresh groceries
Durable groceries
Fuel housing
(oil, gas, electricity)
Telephone units
Fuel (car)
Services
(e.g. hairdresser, trade) (*)
Cars
Travel
Clothes
- 0'·
Auctainer (n
=186)
-1IfII-
Smart Bidder (n = 205)
(*) p:::; 0,10
Figure 9: Future auction goods
Finally the attitude towards the imminent development 0/ private into
commercial consumer auctions was addressed. Both user types judged
this expansion controversially, but Inainly favourably (figure 10). Assortment and choice are expected to improve; brand products are seen
as enriching. Smart bidders do not expect the quality of auctions to be
diminished, while auctainers are much less sure of that. Exploratory
factor analysis showed that all in all auctainers tend to perceive more
chances as weH as more risks than smart bidder, although on both
factors their factor values do not differ significantly.
26
disagree
agree
completely
completely
Brand products enrichment
will use auctions more
danger to auction
more choice
..-
.~.,
less quality *
.
..
IIIIIq
less risk
less friendly
·0"
Auctainer (n = 186)
-II1II-
Smart Bidder (n = 205)
* p:s; 0,05
Figure 10: Attitude towards commercial auctions
5
Discussion
Previous research was not concerned about internet auction customers.
Generally auction shoppers are thought to be thrill-seeking bargain
hunters. Several theories and empirical results indicate the existence
of two different types of auction customers, an experiential type and a
pragmatic type. The empirical study reported here examined behavioral differences between such types. Results offer several implications
for research and for the marketing of internet auctioneers and suppliers.
Foremost research task is the replication and comprehensive validation of shopper types derived in this study. A more thorough integration of auction buying behavior with theories and models of consumer
behavior or the smart shopper phenomenon looks promising.
27
Internet auctioneers should note that experiential shoppers are a minority. In many traits auctainers and smart bidders do not differ much;
auctainers are simply experiential oriented as well. Business models
exploiting this additional utility (auctainment) while neglecting the
core utility of the auction algorithm (individual matchmaking) are Ullable to retain customers in the long run. Because of the peculiar quality of novelty of losing its flavor automatically, this is true even if the
absence of the basic utility remains undetected; it is the more true if
shoppers notice manipulative intentions. The auctainment model thus
seems substantially flawed and the existence of a durable competitive
advantage doubtful. The failure of this approach in the real world must
not be confused with the failure of B2C auctions on the whole. Ebay' s
success in penetrating the market niche left by the auctainment model
demonstrates that this is not true. Future success of auctions will
depend on developing bidder tools that further facilitate transaction
managetnent.
Commercial suppliers who want to put consumer goods and services
up for auction can count on the fact that auctions are a well established, popular, and generally accepted form of online shopping. High
purchase frequency and large percentage of demand covering via auctions make the auction customers studied here a very attractive target
market of e-commerce, one that is highly committed to auctions and
perceives verj few difficulties. Auctainers are somewhat more openminded than smart bidders but they are probably not the better customers: They report more unintentionally high bids and less contentment after the purchase. The lack of return options in most auctions
may explain their tendency not to keep auction goods. Moreover, auctainers must be offered an additional "auction experience", causing
surplus costs to the seller. The biggest obstacle to auction adoption by
sellers probably is the lack of control over progress and final price of
auctions. Judging by their customers, initial chances of consumer
auctions as a marketing tool can be rated excellent.
28
References
Bearchell, C. A. (1999): Will the Internet change your pricing policies?, in:
Marketing Journal, 32 (3), p. 137.
Cassady, R. J. (1967): Auctions and auctioneerinK Berkeley, CA: University of
California Press.
Cronbach, L. J. (1955): Processes affecting scores on "understanding of others" and
"assumed similarity", in: Psychological Bulletin, 52 (3), pp. 177-193.
Elliott, A. C. (2000): Getting started in internet auctions, New York: Wiley.
Esser, B. (1999): Smart Shopping, in: Die Betriebswirtschaft, 59 (6), S. 855-857.
Herschlag, M./Zwick, R. (2000): Internet auctions - a popular and professional
literature review, in: Quarterly Journal of Electronic Commerce, 1 (2), pp.
161-186.
Hoffman, D. L./Novak, T. P. (1996): Marketing in hypermedia computer-mediated
environments: conceptual foundations, in: Journal of Marketing, 60, pp. 5068.
Kotler, P./Armstrong, G. (1994): Principles of Marketing, 6th ed., Englewood
Cliffs, NJ: Prentice Hall.
Li, H./Kuo, C./Russell, M. G. (1999): The impact of perceived channel utilities,
shopping orientations, and demographics on the consumer's online buying
behavior, in: Journal ofComputer-Mediated Communication, 5 (2), pp. 1-20.
Loevenich, P. (2002): Substitutionskonkurrenz durch E-Commerce, Wiesbaden:
Gabler.
Lucking-Reiley, D. (2000): Auctions on the internet: what's being auctioned, and
how, in: The Journal oflndustrial Economics, 48 (3), pp. 227-252.
Lührig, T./Dholakia, N. (2002): OnIine auctions: The emerging new electronic
agora, in: Dholakia, N. et al. (eds.): Global e-commerce and online marketing,
Westport, CT: Quorum, pp. 111-121.
Matthews, S. A. (1987): Comparing auctions for risk averse buyers: A buyer's
point ofview, in: Econometrica, 55 (3), pp. 633-646.
McAfee, R. P./McMillan, J. (1987): Auctions and bidding, in: JournalofEconomic
Literature, 25, pp. 699-738.
McKinsey (2000): All visitors are not created equal, http://www.business2.de/
essentials/studienIMcKinsey_Ecom.pdf [21.4.2001].
Milgrom, P. R./Weber, R. J. (1982): A theory of auctions and competitive bidding,
in: Econometrica, 50 (5), pp. 1089-1122.
Möllenberg, A. (2003): Internet-Auktionen im Marketing aus der Konsumentenperspektive, Braunschweig: Eigenverlag (Doctoral Thesis Technical University of Braunschweig).
Pervin, L. A. (1978): Current controversies and issues in personality, New York,
NY: Wiley.
Porter, M. E. (2001): Strategy and the Internet, in: Harvard Business Review, 23
(3), pp. 63-78.
Schrage, M. (2000): The Creative Mind - To HaI Varian the price is always right,
in: Strategy & Business, 3 (18), pp. 82-89.
29
Smith, C. W. (1989): Auctions: the social construction of value, New York: The
Free Press, MacMillan Inc.
Spohrer, M.lBronold, R. (2000): GfK Online-Monitor, 5. Untersuchungswelle, Juni
2000, Nürnberg: GfK AG Medienforschung.
Tauber, E. M. (1972): Why do people shop, in: Journal of Marketing, 36, pp. 4659.
Vickrey, W. (1961): Counter speculation, auctions, and competitive sealed tenders,
in: Journal ofFinance, 16, pp. 8-37.
Weinhardt, C./Schmidt, C. (2001): Internet-Auktionen, in: Bühner, R. (ed.):
Management-Lexikon, München: Oldenbourg, pp. 383-386.
30
Author
Dr. Antj e Möllenberg was research assistant and lecturer at the Department
of Marketing, Institute of Business Administration and Economics, Technical University of Braunschweig, Germany. She is now a freelance marketing consultant.